Joint cosine transforming and quantizing device and joint inverse quantizing and inverse cosine transforming device

#1. Field of the Invention

#The invention relates to video compression and decompression techniques, more particularly to a video compressing apparatus and a video decompressing apparatus that perform a reduced number of multiplication operations.

#2. Description of the Related Art

#Other features and advantages of the present invention will become apparent in the following detailed description of the preferred embodiments with reference to the accompanying drawings, of which:

#FIG. 1 is a schematic circuit block diagram of a video encoding apparatus that incorporates the preferred embodiments of a joint cosine transforming and quantizing device and a joint inverse quantizing and inverse cosine transforming device according to the present invention;

#FIG. 2 is a schematic circuit block diagram of a video decoding apparatus that incorporates the preferred embodiment of the joint inverse quantizing and inverse cosine transforming device according to the present invention;

#FIG. 1 is a schematic circuit block diagram illustrating a video encoding apparatus that involves discrete cosine transform. A motion estimator 10 receives source input frames and performs motion estimation relative to a previous and/or a future input frame. The motion estimator 10 generates motion vectors in case the input frame is a predictive-coded frame or a bi-directional predictive-coded frame, both of which are commonly referred to as non-intra coded frames. Otherwise, for intra-coded frames, the input frame is provided directly to a joint cosine transforming and quantizing device 20 that includes a discrete cosine transforming (DCT) unit 21 and a quantizer (Q) 22 connected to the DCT unit 21. The DCT unit 21 performs discrete cosine transform on the input frame, and provides transform data to the quantizer 22. The quantizer 22 is responsible for quantizing the transform data from amplitudes that lie within a number of preset ranges to a set of finite levels. The non-intra coded frames are also processed by the joint cosine transforming and quantizing device 20, but only after reference pixel values that come from reference frames stored in a reference frame buffer 30 are subtracted from pixel values of the current frame via a subtracter 11. As such, in case of non-intra coded frames, the joint cosine transforming and quantizing device 20 only processes the differences of the pixel values. A joint inverse quantizing and inverse cosine transforming device 40 includes an inverse quantizer (DeQ) 41 and an inverse discrete cosine transforming (IDCT) unit 42 connected to the inverse quantizer 41. The inverse quantizer 41 performs the inverse operation of the quantizer 22. The IDCT unit 42 performs the inverse operation of the DCT unit 21. Reference frames for the non-intra coded frames are provided by an adder 12. The reference frames can be intra-coded frames that are stored directly in the reference frame buffer 30 after reconstruction by the joint inverse quantizing and inverse cosine transforming device 40. Alternatively, the reference frames can be predictive-coded frames that are further reconstructed with reference to a previous reference frame stored in the reference frame buffer 30. In this case, the adder 12 compensates the IDCT retrieved difference of the pixel values of the predictive-coded frame with its reference frame, and sends the reconstructed predictive-coded frame to the reference frame buffer 30. The motion vector from the motion estimator 10 is used to address the appropriate pixels of the reference frames in the reference frame buffer 30 for conducting the subtraction operation in order to encode non-intra coded frames, and for conducting the compensating operation in order to reconstruct the predictive-coded reference frames. A variable length encoder (VLC) 13 is connected to the output of the quantizer 22, and is used to perform entropy coding, such as Huffman coding, to map the quantized transform data to variable-length codes which are subsequently outputted as compressed video codes.

#The motion estimator 10 removes inter-frame redundancy of the source input frames. The quantizer 22 removes insignificant portions of the video data, which are not sensitive to the human visual system. The variable length encoder 13 further compresses the video data by applying entropy coding techniques so as to result in compressed video data with a compression ratio ranging from tens to hundreds.

#If the source input frames are still images, like photographs, which have no inter-frame relationship, the video encoding apparatus of FIG. 1 can be reduced to comprise only the joint cosine transforming and quantizing device 20 and the variable length encoder 13. The motion estimator 10, the subtracter 11 for providing pixel value differences, the adder 12 for providing reference frames, the reference frame buffer 30 for storing the reference frames, and the joint inverse quantizing and inverse cosine transforming device 40 are omitted.